JurassicPronunciation needs audio and consent
SPK-PRON-INTEL — original coaching practice; evidence before judgement.
Practice only; no IELTS scoring.
Last reviewed:
Observable problem
A listener may report that a particular word was unclear, but a typed answer cannot show how that word sounded. Do not infer pronunciation from spelling or grammar. This page does not collect audio and makes no pronunciation judgement. A useful observation needs a specific word, an utterance anchor and an actual listening context.
Why it happens
A transcript can contain recognition mistakes introduced by a system rather than the speaker. Treat that mismatch as uncertain until a person checks the sound. Pronunciation review needs audio and consent; a consequence-sensitive judgement also needs human review. Audio availability alone is not permission to store, reuse or combine it with another service’s learner records.
Original micro-drill
Original offline micro-drill: choose an invented room label such as “West garden”. Ask a consenting partner to listen to the phrase in “Meet beside the West garden”. The partner repeats what they understood; compare intended and heard meaning without a numeric rating. No recording is necessary, and no audio or typed response is submitted on this page.
New-task transfer check
For a new-task check, use a different fictional location in a complete sentence, such as “The workshop is beside the north gate”. Ask what the listener understood without showing the text first. This is a communication check, not a pronunciation score. For audio-based practice, follow the existing live app and review its consent and privacy information before proceeding.
Next action
Keep the prompt, content cue and any observed evidence separate from assumptions about delivery. Typed answers never produce pronunciation or fluency judgements. Ask permission before recording; no recording or personal information is stored here. Use a genuinely new question for transfer, then choose one content action to practise next. An unclear or missing audio anchor remains insufficient evidence rather than a forced rating.
Observation code: SPK-PRON-INTEL